Language Models with Conformal Factuality Guarantees
CoRR(2024)
摘要
Guaranteeing the correctness and factuality of language model (LM) outputs is
a major open problem. In this work, we propose conformal factuality, a
framework that can ensure high probability correctness guarantees for LMs by
connecting language modeling and conformal prediction. We observe that the
correctness of an LM output is equivalent to an uncertainty quantification
problem, where the uncertainty sets are defined as the entailment set of an
LM's output. Using this connection, we show that conformal prediction in
language models corresponds to a back-off algorithm that provides high
probability correctness guarantees by progressively making LM outputs less
specific (and expanding the associated uncertainty sets). This approach applies
to any black-box LM and requires very few human-annotated samples. Evaluations
of our approach on closed book QA (FActScore, NaturalQuestions) and reasoning
tasks (MATH) show that our approach can provide 80-90
while retaining the majority of the LM's original output.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要